Dynamic Grouping of Web Users Based on Their Web Access Patterns using ART1 Neural Network Clustering Algorithm

TL;DR

This paper introduces an ART1 neural network clustering method to group web users based on their access patterns, demonstrating its effectiveness compared to K-Means and SOM in forming high-quality clusters.

Contribution

The paper presents a novel application of ART1 neural network for web user clustering and compares its performance with established algorithms.

Findings

ART1 achieves higher inter-cluster distance with fewer clusters.

ART1 forms more distinct clusters as the number of clusters increases.

The approach outperforms K-Means and SOM in cluster quality at higher cluster counts.

Abstract

In this paper, we propose ART1 neural network clustering algorithm to group users according to their Web access patterns. We compare the quality of clustering of our ART1 based clustering technique with that of the K-Means and SOM clustering algorithms in terms of inter-cluster and intra-cluster distances. The results show the average inter-cluster distance of ART1 is high compared to K-Means and SOM when there are fewer clusters. As the number of clusters increases, average inter-cluster distance of ART1 is low compared to K-Means and SOM which indicates the high quality of clusters formed by our approach.